Thermoregulatory Correlates of Nausea in Rats and Musk Shrews

www.impactjournals.com/oncotarget/ Oncotarget, Vol. 5, No. 6

Thermoregulatory correlates of nausea in rats and musk shrews

Sukonthar Ngampramuan1, Matteo Cerri2, Flavia Del Vecchio2, Joshua J. Corrigan3, Amornrat Kamphee1, Alexander S. Dragic3, John A. Rudd4, Andrej A. Romanovsky3, and Eugene Nalivaiko5 1 Research Center for Neuroscience and Institute of Molecular Bioscience, Mahidol University, Bangkok, Thailand; 2 Department of Biomedical and Motor Sciences, University of Bologna, Bologna, Italy; 3 FeverLab, Trauma Research, St. Joseph’s Hospital and Medical Center, Phoenix, AZ, USA; 4 School of Biomedical Sciences, Chinese University of Hong Kong, Hong Kong, China; 5 School of Biomedical Sciences and Pharmacy, University of Newcastle, Newcastle, NSW, Australia. Correspondence to: Eugene Nalivaiko, email: [email protected] Correspondence to: Andrej A. Romanovsky, email: [email protected] Keywords: nausea, chemotherapy, temperature, hypothermia. Received: December 21, 2013 Accepted: February 21, 2014 Published: February 22 2014

This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. ABSTRACT: Nausea is a prominent symptom and major cause of complaint for patients receiving anticancer chemo- or radiation therapy. The arsenal of anti-nausea drugs is limited, and their efficacy is questionable. Currently, the development of new compounds with anti-nausea activity is hampered by the lack of physiological correlates of nausea. Physiological correlates are needed because common laboratory rodents lack the vomiting reflex. Furthermore, nausea does not always lead to vomiting. Here, we report the results of studies conducted in four research centers to investigate whether nausea is associated with any specific thermoregulatory symptoms. Two species were studied: the laboratory rat, which has no vomiting reflex, and the house musk shrew (Suncus murinus), which does have a vomiting reflex. In rats, motion sickness was induced by rotating them in their individual cages in the horizontal plane (0.75 Hz, 40 min) and confirmed by reduced food consumption at the onset of dark (active) phase. In 100% of rats tested at three centers, post- rotational sickness was associated with marked (~1.5°C) hypothermia, which was associated with a short-lasting tail-skin vasodilation (skin temperature increased by ~4°C). Pretreatment with ondansetron, a serotonin 5-HT3 receptor antagonist, which is used to treat nausea in patients in chemo- or radiation therapy, attenuated hypothermia by ~30%. In shrews, motion sickness was induced by a cyclical back- and-forth motion (4 cm, 1 Hz, 15 min) and confirmed by the presence of retching and vomiting. In this model, sickness was also accompanied by marked hypothermia (~2°C). Like in rats, the hypothermic response was preceded by transient tail-skin vasodilation. In conclusion, motion sickness is accompanied by hypothermia that involves both autonomic and thermoeffector mechanisms: tail-skin vasodilation and possibly reduction of the interscapular brown adipose tissue activity. These thermoregulatory symptoms may serve as physiological correlates of nausea.

INTRODUCTION either due to cancer itself or secondary to chemotherapy [1]. Many commonly used chemotherapeutic agents Approximately half of cancer patients experience (eg. cyclophosphamide and cisplatin) possess potent nausea and vomiting during the course of their disease, emetogenic properties [2]. Nausea and vomiting produced www.impactjournals.com/oncotarget 1565 Oncotarget by these and other chemothrerapeutic drugs are among the in experimental animals, and we have designed our most severe and feared collateral effects of chemotherapy experiments to elucidate this question. Neither subjective [3, 4]. These side effects not only dramatically worsen perceptions nor sweating can be assessed in conscious quality of life, but may affect the patients’ willingness to rodents. On the other hand, core body temperature, continue their anti-cancer treatment. Of interest, nausea cutaneous temperature, and the preference for ambient has a stronger negative impact on patients’ daily life than temperature are readily measurable, but none have ever vomiting [5]. been used in preclinical studies of nausea. Our principal

The introduction of the 5-HT3 and neurokinin-1 hypothesis was that pro-emetic stimulation causes

(NK1) antagonists substantially improved the management hypothermia and alters behavioral preference for the of chemotherapy-induced vomiting [2, 4]. However, ambient temperature in rodents. paradoxically, their effect on nausea was comparatively modest [6, 7], and it is now becoming recognized that RESULTS nausea and vomiting represent totally different entities, both from the point of view of their pathophysiology and pharmacotherapy [8, 9]. The relative potencies of Provocative motion reduced food consumption in anti-emetic drugs to differentially reduce nausea and rats. emesis were not immediately recognized as nausea was only a secondary end-point in most clinical studies. Likewise, most recent pre-clinical studies of novel anti- Provocative motion had substantial and significant emetic substances are based on emetic responses, and it anorexic effect (Fig. 1): during 30-min test period, rats is assumed that drugs which suppress retching/vomiting subjected to rotation consumed 42% less food compared to in animals may be equally good in suppressing nausea controls (1.4±0.1 vs. 2.5±0.3g, respectively; p<0.05); this in humans. As noted, this is not the case. The situation effect was even larger (49%, or 0.43±0.02 vs. 0.84±0.05 g, is even more complicated in ‘nausea’ studies performed respectively; p<0.05) if food consumption was assessed as in common laboratory animals such as rats and mice, a percentage of increase in body weight. In contrast, there as these animal do not have emetic reflex. Here, a was no effect of rotation on water consumption (2.6±0.3 researcher’s choice for assessing nausea is limited to vs. 2.7±0.3, or 0.85±0.03 vs. 0.85±0.05%, respectively; just a few biological indices with either poor temporal p>0.05 for both). resolution (e.g. reduced locomotion or food intake) or questionable face validity (e.g. pica - kaolin consumption) Provocative motion caused hypothermia in rats. and specificity (eg. conditioned food aversion) (reviewed in [8]). During the initial (basal) thermogradient session, This differential action of anti-emetic drugs on there was no difference in basal core body temperature nausea and emesis in the clinic led to the realization that between animals exposed to provocative motion and there may be different pathways and control systems for controls. In all tested rats, rotation caused consistent falls nausea and vomiting [8]. Furthermore, it is now clear in the core body temperature (Fig. 2A). These falls became that the wealth of existing mechanistic knowledge on apparent within the first 5-7 min of rotation, and did not the neural mechanism of the vomiting reflex, with its circuitry located mainly in the brainstem [8, 10], is not directly applicable to nausea – a subjective experience A. Food consumption. B. Water consumption. N = 9 that presumably originates in some forebrain areas. (each group) Consequently, identification of the key transmitter systems and receptors involved in nausea perception is essential for the improved care of the cancer patients. Providing that major methodological obstacle for advancing in this field * * is lack of adequate animal models for studying nausea [11], development and validation of new potential physiological indices for such studies is of major importance. In humans, nausea is commonly associated with facial pallor, sweating and gastric awareness [8]. It is less known, but well documented that in humans nausea is also associated with profound disturbances in thermoregulation, Fig 1: Provocative motion substantially reduced including falls in the core body temperature, modified food consumption (A) but had no effect on water perception of the ambient temperature and preference consumption (B) in rats. Both food and water consumption for cooler environment [12-15]. It is currently unknown were determined in absolute values and as a percentage of body whether pro-emetic stimulation elicits similar disturbances weight. White bars - control group; grey bars - after provocative motion. * - significantly different from control (P<0.01). www.impactjournals.com/oncotarget 1566 Oncotarget reach steady state or reverse until rotation was terminated. motion, we observed rises in core body temperature Following return to the thermogradient, core body in rats returned to their home cages after the initial temperature returned to the basal level within 30 min. In thermogradient session. Their temperature started to fall contrast to hypothermic responses elicited by provocative after reaching peak, but slightly increased again following

Fig 2: Provocative motion elicited hypothermic responses (A) and had no effect on the preference for the ambient temperature (B) in rats. Note that in control animal, moving them to home cage and back to the thermogradient setup provoked rises in body temperature. ** - different from basal values, p<0/01.

Fig 3: Provocative motion causes rise in tail temperature indicative of vasodilatation. A - mean group data (n=7) from the experiment where either ondansetron (0.5 mg/kg) or vehicle were administered i.p. 30 min prior to the onset of rotation. B - mean group data from the control (without rotation) experiment conducted in the same animals. Statistical comparison was made between data obtained during 5 min just prior to the onset of rotation and during maximal tail warming (A) or between corresponding data points (B). * - different from pre-rotation values. Panels C and D represent infrared images (presented in pseudo-colour scale) of a rat before and during rotational stimulation. Note the substantial increase in the temperature of the tail. Arrows indicate tail region used for data collection. www.impactjournals.com/oncotarget 1567 Oncotarget Fig 4: 5-HT3 receptor blockade slightly reduces hypothermia induced by provocative motion. A - mean group data from the experiment where either ondansetron (0.5 mg/kg) or vehicle were administered i.p. 30 min prior to the onset of rotation. B - mean group data from the control (without rotation) experiment conducted in the same animals. Statistical comparison was made between data obtained during 5 min just prior to the onset of rotation and during the last 5 min of rotation (A) or between corresponding data points (B). ** - different from pre-rotation values (p<0.01).

Fig 5: Provocative motion causes rise in tail temperature associate with falls in the interscapular and lumbar temperature in the house shrew (S. murinus). A – mean group values for changes in the temperature of interscapular (black circles), lumbar (triangles) and tail (squares) regions and the difference between interscapular and lumbar temperatures (white circles). Statistical comparison was made between data obtained during 2 min just prior to the onset of rotation and 2 min during maximal tail warming. * and ** - different from pre-stimulation, p<0.05 and p<0.01, respectively. Panels B and C represent infrared images (presented in pseudo-colour scale) of a shrew before and during rotational stimulation.. Arrows indicate tail region used for data collection. www.impactjournals.com/oncotarget 1568 Oncotarget the return back to the thermogradient. Data values for the to fall as the the tail temperature rose. These results are core body temperature changes and results of statistical shown in Fig. 5A, with data values presented near the tests are presented in Fig. 2A.There were no differences traces. Tail temperature changes were relatively fast, with in preferred ambient T between control and experimental onset within 1 min of stimulation and with peak at 172±12 conditions (Fig. 2B). Provocative motion had no effect on s (range 120-240 s). These effects were observed in six of animals’ preferred ambient T for more than 2 h. seven animals tested; of note, the non-responder had lower values of basal interscapular, lumbar and tail temperature Provocative motion causes an increase in the tail (36.2, 33.6 and 20.2°C, respectively) compared to five temperature in rats. other shrews. An example of infrared images obtained in one shrew before and 6 min after the onset of rotation are presented in Fig. 5B and C. Analysis of infrared images revealed that within 10 minutes from the onset of rotation, tail temperature started DISCUSSION to rise, reached peak at about 20 min, and then gradually returned to the baseline (Fig. 3A). Ondansetron had no Our major novel finding is that in rats, provocative effect on the amplitude of tail warming (p>0.05). In the motion causes anorexic effect indicative of nausea state control experiment (without rotation), neither Ringer nor associated with a rapid and robust fall in the core body ondansetron affected tail temperature (Fig. 3B). Mean data temperature; and that this latter effect is mediated, at values are also presented in Fig. 3. An example of infrared least in part, by vasodilatation in the thermoregulatory images obtained in one rat before and 20 min after the tail vascular bed. This finding was confirmed in the musk onset of rotation are presented in Fig. 3C and D. shrew, where tail vasodilation preceded the first vomiting episode, when animals likely experienced nausea. Our Ondansetron attenuates hypothermic responses to observations mirror thermoregulatory changes reported provocative motion in rats. in humans during nausea [12-15]. Together, they form the basis for a new concept – that disturbances in In this group of rats (n=9), provocative motion thermoregulation is a core element in the pathophysiology elicited hypothermic responses that were quantitatively of nausea. Indeed, looking from this angle at the symptoms and qualitatively similar to those reported in Exp. 2 above. that accompany nausea in humans, one would immediately Pre-treatment with ondansetron partially but significantly realize that they form a “thermoregulatory cluster”. Firstly, reduced this rotation-induced hypothermia (Fig. 4A), with and somewhat surprisingly, nausea-related sweating the value of temperature fall post-drug (-1.3±0.2°C) being has never been considered as an indication of altered smaller compared to the post-vehicle value (-1.8±0.2°C; temperature control; meanwhile, it is known that nausea p<0.05). Ondansetron alone did not have any effect on the does trigger not only “stress-related” palmar sweating, core body temperature (Fig. 4B). but also thermoregulatory sweating on the dorsum of the hand [16]. Secondly, it is now documented that in Provocative motion affects surface body humans experiencing nausea, subjective perception of temperature in musk shrews. the ambient temperature as well as subjective preference for it are altered [13]. Lastly, as it was earlier found in humans [12, 14, 15], and as we now report in two different Our initial intention was to record core body mammalian species, pro-emetic stimuli appear to suppress temperature in shrews by means of biotelemetry, similar cold defence, by causing cutaneous vasodilatation leading to the rat study. Unfortunately it appeared that shrews did to heat dissipation and fall in the core body temperature. not tolerate intraperitonial implantation of the transmitters, resulting in high mortality. For this reason we report Choice of the pro-emetic intervention. here only effects of provocative motion on the surface temperature. Similar to previous reports, motion stimuli caused repetitive retching/vomiting episodes in all tested It is well established that sensitivity to animals (n=7). The mean number of these episodes was chemotherapy-induced nausea is higher in those patients 5.7±0.7 (range 4-8 in 15 min), with a mean latency of who are also susceptible to motion sickness and/or 195±41 s (range 90-380 s). Analysis of infrared images with a history of nausea during pregnancy [17, 18]. revealed consistent and substantial reduction in the This suggests that sensitivity to nausea is determined at cutaneous temperature in the interscapular and lumbar integrative centers in the brain receiving different inputs. areas during provocative motion; this was preceded by a In turn, this allows measuring the sensitivity to one type transient increase in the tail temperature that subsequently of emetic stimuli to assess sensitivity to another. This fell below the basal level. Of note, the difference between is of major importance for our study as the most direct the interscapular and the lumbar temperatures tended approach here would be to subject the experimental www.impactjournals.com/oncotarget 1569 Oncotarget animals to chemotherapeutic agents only. However, the with an increase in the tail blood flow. multiple biological actions of these drugs (e.g. involving Similarity between temperature changes elicited by gross damage to tissues) and the complexity of relevant provocative motion in humans, rats and shrews suggests pathways (e.g. both peripheral nerve input and input that these mammalian species possibly possess similar from the area postrema) renders great difficulty in tracing neural mechanisms linking nausea to thermoregulation. causative links between their initial sites of action and Consequently, thermoregulatory disturbances may not our end-point measurements. For this reason, we induced only be considered as a key accompanying symptom a ‘nausea-like’ state by provocative motion – a “clean” of nausea, but in fact could be used as a valuable stimulus, acting through well characterized vestibular physiological correlate of this disorder in experimental pathways, and not involving biochemical effects of animals. This is supported by our finding that falls in body chemotherapy. Providing that nausea-related effects of temperature preceded retching/vomiting in shrews: this chemotherapy are long-lasting (days), relatively fast means that if they experienced sensation akin to nausea recovery from provocative motion is another obvious prior to these episodes, it was associated with altered advantage in addressing mechanisms of nausea. temperature control. Of course our proposal requires further validation, in first instance by testing whether Body core and tail temperature as a new observed effects on the temperature are sensitive to drugs physiological correlate of nausea in animals. used to treat nausea. In the current study, we started this validation from ondansetron, a 5-HT3 receptor antagonist with proven antiemetic potency [29]. The drug only Provocative motion has been used previously to moderately attenuated motion-induced hypothermia in examine signs of motion sickness in both musk shrews and our rats. It is unlikely that this poor effect was due to the rats. In shrews, motion stimuli reliably provoked retching low dose as we have used a dosage validated in other and vomiting [19, 20], similarly to our observation. With in vivo bioassays [30]. Rather, if hypothermia reflects a the exception of a single article that describes motion- nausea-like state as we suggest, such a poor effect should induced alterations in gastric myeloelectric activity [21], be expected, providing a recognized dissociation between all other studies conducted on shrews did not quantify any anti-nausea and antiemetic potency of the drug in humans other indices except emetic episodes. Our work is thus the [29]. Our control experiments (without rotation) confirmed first to report motion-induced temperature-based effects that ondansetron alone does not have any effect on the in this species. Since in humans nausea usually precedes core body temperature or on the tail temperature. The vomiting, we consider the fact that thermoregulatory former observation is in accord with the earlier studies that changes preceded vomiting/retching episodes in shrews address this issue [31, 32]. as an indirect evidence for a nausea-like state in these Our negative thermogradient data indicate that animals. contrarily to humans, pro-emetic stimulation does not Reduced food consumption following provocative change rats’ preference for the ambient temperature. The motion has been reported previously in rats [22], and our reason of this discrepancy is currently unknown, as our results are in accord with these earlier data. However, thermogradient methodology is highly sensitive and well most provocative motion studies conducted in rats validated [33]. It may be that in rats the change in the were focussed on quantifying pica – an unconventional ambient temperature preference does occur, but quickly consumption of kaolin (eg. [23, 24]. While having some restores after termination of the stimulus. Presentation of predictive validity, this approach has relatively low pro-emetic stimuli in the thermogradient may clarify this temporal resolution; morepver, nausea-related pica is issue in future experiments. absent not only in humans but also in musk shrews and mice [25, 26]. In addition to pica, in rats provocative Thermoeffectors mediating motion-induced motion reduces locomotor activity [27] and provokes effects on body temperature. conditioned taste aversion [28]. There is only one study where the effect of provocative motion on body temperature was examined in rats: Ossenkopp at al. A rise in the tail temperature was associated with [27] reported that rectal temperature fell by about 20C in falls in body temperature in both rats and shrews. This animals subjected to horizontal rotation (70 rpm for 30- suggests that in both species, dilatation of the tail vascular min). Our work not only reconfirms these earlier finding bed represented the mechanism underlying heat loss but also takes it further, by: i) non-invasive measurement and leading to hypothermia. It appears however that of core temperature that eliminate the confounding of this was not the sole cause of hypothermia observed the stress response elicited by rectal measurement; ii) in our experiments. Lack of ondansetron effect on the providing a real-time temporal course of the temperature motion-induced rise in the tail temperature suggests that fall; and iii) demonstrating that hypothermia is associated the anti-hypothermic action of the drug was not due to

www.impactjournals.com/oncotarget 1570 Oncotarget preventing tail vasodilation. In rodents, temperature responsible for motion sickness could interfere with balance is achieved by coordinated regulation of heat loss the descending thermoregulatory control. Changes via cutaneous blood vessels and heat generation by the in subjective perception of ambient temperature and brown adipose tissue (BAT) (see [34, 35] for reviews). It preference for a cooler environment, commonly known is thus possible that motion-induced hypothermia was a and recently documented effects of motion sickness in result of two synergistic effects - increase in the tail blood humans [13], suggest that this interference occurs quite flow (i.e. increased heat dissipation) and reduction in the high in the neuraxis, potentially in the preoptic anrerior BAT thermogenesis, and that ondansetron affected only hypothalamus. the latter. The idea of downregulation of the BAT activity Knowledge about the pathways controlling tail by provocative motion is supported, at least in shrews, vascular tone and BAT does not bring any clarity in the by observation that not only did the provocative motion question why ondansetron had differential effects on reduce interscapular and lumbar skin temperatures, but it these two thermoeffectors. There is currently no data also tended to reduce the temperature difference between on involvement of 5-HT3 signalling in either of these these two regions, which is indicative of reduced BAT pathways. There is however an additional possibility thermogenesis [36]. as it is known that provocative motion, similar to other Our earlier analysis of thermoeffector threshold emetic stimuli, reduces locomotion in rats [27]. If data [37] suggests that deep hypothermic responses ondansetron in our study attenuated this effect, it could (such as the ones reported here) occur primarily due to be that locomotion-generated heat resulted in reduced a drastic decrease in the threshold body temperature for hypothermia during motion stimulation in animals BAT thermogenesis (with a much smaller and inconsistent treated with the drug. Unfortunately telemetric signal for reduction in the threshold temperature for skin locomotion cannot be recorded during rotation, and we are vasodilation), thus leading to poikilothermy. The latter, in unable to confirm or disapprove this possibility. turn, results in hypothermia – if the ambient temperature is The most challenging task of our discussion is low enough. The best studied response of this type is LPS- to provide suggestions about the site of origin of the induced hypothermia [36]. Since bacterial endotoxemia final neural input that provokes perturbations in the is frequently associated with nausea and vomiting (e.g. thermoregulatory control. Currently, the dominating [38, 39]), it is tempting to suggest that LPS hypothermia theory of motion sickness is the one of sensory conflict; and hypothermia induced by pro-emetic stimulation it postulates that when converging vesibular, visual and may involve similar mechanisms. It would thus be most proprioceptive input patterns differ from the expected interesting to determine the effect of provocative motion sensory pattern stored in memory, spatial orientation is on thermoeffector thresholds. disturbed, and this leads to motion sickness [40]. Until recently, there was no direct data regarding potential Neural pathways mediating hypothermia elicited anatomical location of neurons generating “sensory by the provocative motion. mismatch” signal. In an elegant studies conducted in rats [41] demonstrated that this might happen in the hippocampus, whose role in spatial orientation and Rotation stimuli used in the present work caused learning is well recognized. The first human brain imaging alterations in vestibular and possibly visual sensory inputs, study of motion sickness revealed several regions that and there must be some area within the brain where this were activated during the experience of nausea, namely sensory information was converted into the command the medial prefrontal cortex, the insular cortex, the signals targeting thermoeffectors. Location of central amygdala, the putamen and the locus coeruleus [42]; command neurons and efferent pathways controlling somewhat surprisingly, hippocampus was not among thermoeffectors is relatively well understood, at least them. Clearly, cited works provide a good framework for in rats [34, 35]. Information from central (brain) and future experimental determination of motion sickness- peripheral thermosensors is integrated in the preoptic activated inputs to thermoregulatory pathways. anterior hypothalamus that sends excitatory projection to the dorsomedial hypothalamus, a major integrative Physiological significance and perspectives. centre for autonomic output. Descending presympathetic pathways from the DMH relay in the periaqueductal grey and then in the medullary raphe/parapyramidal Provocative motion could be considered as a area and in the spinal cord, where separate populations stressful intervention, and in this context hypothermia of sympathetic neurones control two thermoeffectors caused by it seems paradoxical providing all other – brown adipose tissue responsible for non-shivering stressors (restraint, social interaction, novelty, footshocks thermogenesis and cutaneous vascular bed responsible for etc.) provoke hyperthermic responses in rodents. heat dissipation/conservation. It is thus clear that there is a Ossenkopp [27] presented an interesting explanation of limited number of neural targets where neural mechanisms this apparent paradox. Based on the idea that lowing body temperature could represent a defence mechanism against www.impactjournals.com/oncotarget 1571 Oncotarget intoxication and on the Treisman’s theory that aversive University of Bologna (Italy). All rats were adult males sensation of nausea which accompanies motion sickness of Wistar strain, weighing 220-270 g. Experiment 5 was is a by-product of evolutionary developed defence against conducted on the adult male house musk shrews (Suncus ingested neurotoxins [43], authors postulated that motion murinus) at the Chinese University of Hong Kong. sickness-induced hypothermia is a manifestation of this In Experiment 1, we studied the effect of evolutionary beneficial reaction. Its high evolutionary provocative motion on food and water consumption in relevance might explain the fact that provocative motion rats. All food and water was removed from the animals’ not only led to a fall in the abdominal temperature cages at the beginning of the light cycle (6:00 AM), and compared to baseline, but in addition counteracted animal weights were recorded. Rats were either rotated hyperthermic responses provoked by animal handling on a turntable at 0.75 Hz (experimental group, n=9) or (Fig. 2A). not rotated (control group, n=9) for 40 min just prior to As we stated in the Introduction, the major obstacle the start of the dark cycle (6:00 PM). Animals were then in preclinical studies of nausea is the lack of ability to returned to their home cages where pre-weighed water and assess its cardinal symptoms (subjective experience and standard rat chow were made available to them. Food and sweating) in experimental animals. The only established water were then weighed 30 min after the onset of the dark biochemical marker of nausea in humans, elevated plasma cycle. vasopressin [44, 45], has been confirmed in cats [46] and In Experiment 2, we studied the effect of ferrets [47] but not in rats [44, 48]. Consequently, rodent provocative motion on the core body temperature and on

studies of motion sickness-induced nausea have to rely preferred ambient temperature (Tamb) in another group of on indirect indices, often with poor temporal resolution rats. Each animal’s body temperature was recorded using (eg. locomotor activity, food consumption) or limited an implantable datalogger (SubCue, Calgary, Alberta, face validity (eg. pica that, in addition, is species specific Canada), which was stitched to the inner abdominal [25, 26]). Quantifying retching or vomiting in species wall via midline laparotomy under ketamine-xylazine- possessing emetic reflex (eg. shrews) now appears not to acepromazine anesthesia (55.6, 5.5, and 1.1 mg/kg i.p. be an ideal solution for nausea studies as here the major respectively) and antibiotic protection (enrofloxacin, 5 difficulty is in the totally different neural substrates mg/kg). Animals were allowed to recover for 3 days prior responsible for nausea and for vomiting. It is most likely to experimentation. On the day of the experiment, rats that relying on emetic responses in preclinical studies were placed inside a thermogradient apparatus (previously

is the major reason for pharmacological dissociation of described by [33]) to assess baseline preferred Tamb. The current anti-emetics that efficiently suppress vomiting but thermogradient apparatus consisted of six 217-cm-long are less potent in preventing nausea [9, 29]. aluminum channels that run between two water tanks. We believe that our current findings, in combination Two electric heating units (PolyScience, Niles, IL, USA)

with the solid evidence obtained in humans and discussed heated the water inside one tank to maintain a Tamb of above, represent a firm basis for the claim that altered 30oC inside the channels at the “warm” end, whereas the thermoregulation is a core pathophysiological element of water tank at the “cold” end was constantly perfused with nausea in mammals. Consequently, assessing temperature- 10% ethylene glycol by an external-circulation cooling o related indices in experimental animals subjected to pro- pump (PolyScience) to maintain a Tamb of 15 C inside the emetic stimuli, including chemotherapeutic agents, may channels at the opposite end. By manipulating the water represent a promising novel approach for determining temperatures at either end, a near-linear temperature brain neural circuits responsible for nausea, and for gradient of 0.07oC/cm was maintained along each of the

assessing its pharmacological sensitivity. six channels. Tamb in the thermogradient was continuously monitored via five evenly-spaced (50cm) thermocouples METHODS per channel. The thermocouples were fed into a TempScan 1000A receiver (Omega, Stamford, CT, USA), which relayed temperature data to the TempWindows Animals and experimental protocols. software program (Omega) in real-time. A high-definition video camera (Panasonic WV-CP284, Panasonic, Kadoma, Osaka, Japan), positioned directly overhead, captured each Investigation has been conducted in accordance animal’s location inside the thermogradient and relayed it with the ethical standards and according to the to the TopScan software (Cleversys, Reston, VA, USA) Declaration of Helsinki and according to national and for tracking along the established temperature gradient. international guidelines and has been approved by the Rats were then removed, individually housed in their authors’ institutional review board. Experiments 1 and home cages, and placed on a turntable. The turntable 2 were conducted at St. Joseph’s Hospital and Medical was rotated at 0.75 Hz (experimental group, n=15) or not Center (Phoenix, AZ, USA); Experiment 3 – at Mahidol rotated (control group, n=15) for 40 min. Immediately University (Bangkok, Thailand) and Experiment 4 - at the thereafter, rats were returned to the thermogradient www.impactjournals.com/oncotarget 1572 Oncotarget apparatus for 3 hours. All animals were well-adapted to used in assessing the preferable ambient temperature. the experimental setups prior to experimentation. No Radiotelemetric temperature data (Exp. 3) was sampled drugs were administered during this experiment. at 1 Hz and acquired using MacLab-8s data acquisition In Experiment 3 we studied the effects of system (ADInstruments, Sydney, Australia), and then provocative motion on the tail temperature in reduced by averaging within every 2-min interval. uninstrumented rats. The protocol (recording timing, Infrared images from rats (Exp. 4) were collected from drug injections and rotation parameters) was identical to individual frames at 2-min intervals; tail temperature Experiment 3. For assessing tail temperature before and was assessed at the distal region of the proximal third during provocative motion, infrared video images were of the tail. Infrared images from shrews (Exp. 5) were acquired using a thermoimager (FLIR-A20, Flir Systems, also collected from individual frames at 2-min intervals. Wilsonville, OR, USA). There were 9 rats in this group. Using proprietary SmartIRMNet software (Keii, China), In experiment 4, we tested whether hypothermic we then determined the superficial skin temperature from effect of provocative motion is sensitive to antiemetic three areas: interscapular region, bottom of the trunk at drug ondansetron. Telemetric transmitters (TA11CA-F40, the midline, and tail area 1 cm distal from the base. Both Data Sciences International, USA) were aseptically cameras have built-in self-calibration capabilities. implanted into the peritoneal cavity under isoflurane (2% For statistical comparison, data were selected

in O2) anaesthesia, and animals were left to recover for as following: Exp. 2, body T – mean of the two last 1 week; they were housed individually during this time. points just before rotation vs. mean of the two points at On the day of experiment, transmitters were turned on, maximum (for control animals) or at minimum (rotated home cages were placed on a turntable that was stationary, animals); Exp. 2, thermogradient - mean of the two last and baseline recording of the core body temperature was points just before rotation vs. mean of the two first points obtained for 1 h. At the middle of this period (i.e. at 30 post-rotation; Exp. 3, rotation condition - mean of the min), animals received i.p. injection of either antiemetic four last points just before rotation vs. mean of the two ondansetron (0.5 mg/kg in 0.5 ml Ringer) or vehicle four points during maximal rise in the tail T; Exp. 3, (0.5 ml Ringer). Thirty minutes later, the turntable was control – corresponding time intervals; Exp. 4, rotation switched on, and rotation was maintained for 40 min at condition - mean of the four last points just before rotation 0.75 Hz, This protocol was repeated one week later, with a vs. mean of the two four points during maximal fall in counter-balanced administration of ondansetron or vehicle. the abdominal; Exp. 4, control – corresponding time During subsequent week 3 and 4, we determined whether intervals; Exp. 5 mean of the two last points just before there were any effects of the drug or vehicle alone on the rotation vs. mean of the two four points during maximal temperature. For this purpose, the protocol was repeated rise or fall in each parameter. Statistical significance for again once a week, but without rotation. There were 9 the differences in water consumtion was performed usin animals in this experiment. Ondansetron was from Sigma unpaired Student’s t-test. Statistical differences in the (USA). core body and the cutaneous temperatures were assessed In Experiment 5, we studied effects of provocative by means of two-way ANOVA (treatment vs. time) with motion on the surface body and tail temperature in adult Bonferroni correction, with p<0.05 being the threshold for house musk shrews (Suncus murinus – an insectivore significance. possessing the vomiting reflex; N = 7). Animals were housed in groups of 4 per cage. On the day of experiment, REFERENCES: each animal was moved to a new clean cage and remained in it for 40 min for habituation prior to recording. A cage 1. Warr DG. Chemotherapy-and cancer-related nausea and with an animal was placed on a laboratory shaker, with vomiting. Current oncology. 2008;15:S4-9. an infrared camera (KC500, Keii, China) fixed above the 2. Di Maio M, Bria E, Banna GL, Puglisi F, Garassino MC, cage. A 15-min basal recording was followed by 15-min of Lorusso D and Perrone F. Prevention of chemotherapy- provocative motion (1 Hz, 4 cm linear displacement). We induced nausea and vomiting and the role of neurokinin also quantified the number of retching/vomiting episodes 1 inhibitors: from guidelines to clinical practice in solid and their latencies. We did not administer drugs in this tumors. Anti-cancer drugs. 2013;24:99-111. experiment. The ambient temperature for all experiments 3. Lindley C, McCune JS, Thomason TE, Lauder D, Sauls A, was 20-21oC. Adkins S and Sawyer WT. Perception of chemotherapy side effects cancer versus noncancer patients. Cancer practice. Data acquisition, analysis and statistical 1999;7:59-65. evaluation 4. de Boer-Dennert M, de Wit R, Schmitz PI, Djontono J, v Beurden V, Stoter G and Verweij J. Patient perceptions of The dataloggers (Exp. 2) sampled core body the side-effects of chemotherapy: the influence of 5HT3 temperature every 3 min; a similar sampling rate was antagonists. British journal of cancer. 1997;76:1055-61.

www.impactjournals.com/oncotarget 1573 Oncotarget 5. Bloechl-Daum B, Deuson RR, Mavros P, Hansen M and experimental model for motion sickness. Life sciences. Herrstedt J. Delayed nausea and vomiting continue to 1988;43:413-20. reduce patients’ quality of life after highly and moderately 20. Rudd JA, Ngan MP and Wai MK. Inhibition of emesis by emetogenic chemotherapy despite antiemetic treatment. tachykinin NK1 receptor antagonists in Suncus murinus Journal of clinical oncology : official journal of the (house musk shrew). European journal of pharmacology. American Society of Clinical Oncology. 2006;24:4472-8. 1999;366:243-52. 6. Foubert J and Vaessen G. Nausea: the neglected symptom? 21. Percie du Sert N, Chu KM, Wai MK, Rudd JA and Andrews European journal of oncology nursing : the official journal PL. Telemetry in a motion-sickness model implicates the of European Oncology Nursing Society. 2005;9:21-32. abdominal vagus in motion-induced gastric dysrhythmia. 7. Navari RM. Management of chemotherapy-induced nausea Experimental physiology. 2010;95:768-73. and vomiting : focus on newer agents and new uses for 22. Riccio DC and Thach JS, Jr. Response suppression older agents. Drugs. 2013;73:249-62. produced by vestibular stimulation in the rat. Journal of the 8. Stern R, Koch K and Andrews P. Nausea. New York: experimental analysis of behavior. 1968;11:479-88. Oxford University Press; 2011. 23. Mitchell D, Laycock JD and Stephens WF. Motion 9. ASHP. ASHP Therapeutic Guidelines on the Pharmacologic sickness-induced pica in the rat. The American journal of Management of Nausea and Vomiting in Adult and Pediatric clinical nutrition. 1977;30:147-50. Patients Receiving Chemotherapy or Radiation Therapy or 24. McCaffrey RJ. Appropriateness of kaolin consumption Undergoing Surgery. American journal of health-system as an index of motion sickness in the rat. Physiology & pharmacy : AJHP : official journal of the American Society behavior. 1985;35:151-6. of Health-System Pharmacists. 1999;56:729-64. 25. Yamamoto K, Ngan MP, Takeda N, Yamatodani A and 10. Hornby PJ. Central neurocircuitry associated with emesis. Rudd JA. Differential activity of drugs to induce emesis The American journal of medicine. 2001;111 Suppl and pica behavior in Suncus murinus (house musk shrew) 8A:106S-112S. and rats. Physiology & behavior. 2004;83:151-6. 11. Sanger GJ and Andrews PL. Treatment of nausea and 26. Liu YL, Malik N, Sanger GJ, Friedman MI and Andrews vomiting: gaps in our knowledge. Autonomic neuroscience PL. Pica--a model of nausea? Species differences in : basic & clinical. 2006;129:3-16. response to cisplatin. Physiology & behavior. 2005;85:271- 12. Nobel G, Eiken O, Tribukait A, Kolegard R and Mekjavic 7. IB. Motion sickness increases the risk of accidental 27. Ossenkopp KP, Rabi YJ, Eckel LA and Hargreaves EL. hypothermia. European journal of applied physiology. Reductions in body temperature and spontaneous activity 2006;98:48-55. in rats exposed to horizontal rotation: abolition following 13. Nobel G, Tribukait A, Mekjavic IB and Eiken O. Effects chemical labyrinthectomy. Physiology & behavior. of motion sickness on thermoregulatory responses in a 1994;56:319-24. thermoneutral air environment. European journal of applied 28. Ossenkopp KP. Area postrema lesions in rats enhance physiology. 2012;112:1717-23. the magnitude of body rotation-induced conditioned taste 14. Mekjavic IB, Tipton MJ, Gennser M and Eiken O. Motion aversions. Behavioral and neural biology. 1983;38:82-96. sickness potentiates core cooling during immersion in 29. Olver IN. Update on anti-emetics for chemotherapy-induced humans. The Journal of physiology. 2001;535:619-23. emesis. Internal medicine journal. 2005;35:478-81. 15. Cheung B, Nakashima AM and Hofer KD. Various anti- 30. Tuerke KJ, Winters BD and Parker LA. Ondansetron motion sickness drugs and core body temperature changes. interferes with unconditioned lying-on belly and acquisition Aviation, space, and environmental medicine. 2011;82:409- of conditioned gaping induced by LiCl as models of 15. nausea-induced behaviors in rats. Physiology & behavior. 16. McClure JA, Fregly AR, Molina E and Graybiel A. 2012;105:856-60. Response from arousal and thermal sweat areas during 31. Kandasamy SB. Effect of ondansetron and ICS 205-930 on motion sickness. Aerospace medicine. 1972;43:176-9. radiation-induced hypothermia in rats. Radiation research. 17. Morrow GR. The effect of a susceptibility to motion 1997;147:741-6. sickness on the side effects of cancer chemotherapy. 32. Mazzola-Pomietto P, Aulakh CS and Murphy DL. Cancer. 1985;55:2766-70. Temperature, food intake, and locomotor activity effects 18. Roscoe JA, Bushunow P, Morrow GR, Hickok JT, Kuebler of a 5-HT3 receptor agonist and two 5-HT3 receptor PJ, Jacobs A and Banerjee TK. Patient expectation is a antagonists in rats. Psychopharmacology. 1995;121:488-93. strong predictor of severe nausea after chemotherapy: a 33. Almeida MC, Steiner AA, Branco LG and Romanovsky University of Rochester Community Clinical Oncology AA. Neural substrate of cold-seeking behavior in endotoxin Program study of patients with breast carcinoma. Cancer. shock. PloS one. 2006;1:e1. 2004;101:2701-8. 34. Romanovsky AA. Thermoregulation: some concepts have 19. Ueno S, Matsuki N and Saito H. Suncus murinus as a new changed. Functional architecture of the thermoregulatory www.impactjournals.com/oncotarget 1574 Oncotarget system. American journal of physiology Regulatory, integrative and comparative physiology. 2007;292:R37-46. 35. Morrison SF and Nakamura K. Central neural pathways for thermoregulation. Frontiers in bioscience. 2011;16:74-104. 36. Marks A, Vianna DM and Carrive P. Nonshivering thermogenesis without interscapular brown adipose tissue involvement during conditioned fear in the rat. American journal of physiology Regulatory, integrative and comparative physiology. 2009;296:R1239-47. 37. Romanovsky AA. Do fever and anapyrexia exist? Analysis of set point-based definitions. American journal of physiology Regulatory, integrative and comparative physiology. 2004;287:R992-5. 38. Todd JK. Staphylococcal toxin syndromes. Annual review of medicine. 1985;36:337-47. 39. Brock-Utne JG and Gaffin SL. Endotoxins and anti- endotoxins (their relevance to the anaesthetist and the intensive care specialist). Anaesthesia and intensive care. 1989;17:49-55. 40. Reason J and Brand J. Motion sickness. London: Academic Press; 1975. 41. Aitake M, Hori E, Matsumoto J, Umeno K, Fukuda M, Ono T and Nishijo H. Sensory mismatch induces autonomic responses associated with hippocampal theta waves in rats. Behavioural brain research. 2011;220:244-53. 42. Napadow V, Sheehan JD, Kim J, Lacount LT, Park K, Kaptchuk TJ, Rosen BR and Kuo B. The brain circuitry underlying the temporal evolution of nausea in humans. Cerebral cortex. 2013;23:806-13. 43. Treisman M. Motion sickness: an evolutionary hypothesis. Science. 1977;197:493-5. 44. Rowe JW, Shelton RL, Helderman JH, Vestal RE and Robertson GL. Influence of the emetic reflex on vasopressin release in man. Kidney international. 1979;16:729-35. 45. Fisher RD, Rentschler RE, Nelson JC, Godfrey TE and Wilbur DW. Elevation of plasma antidiuretic hormones (ADH) associated with chemotherapy-induced emesis in man. Cancer treatment reports. 1982;66:25-9. 46. Fox RA, Keil LC, Daunton NG, Crampton GH and Lucot J. Vasopressin and motion sickness in cats. Aviation, space, and environmental medicine. 1987;58:A143-7. 47. Hawthorn J, Andrews PL, Ang VT and Jenkins JS. Differential release of vasopressin and oxytocin in response to abdominal vagal afferent stimulation or apomorphine in the ferret. Brain research. 1988;438:193-8. 48. Verbalis JG, McHale CM, Gardiner TW and Stricker EM. Oxytocin and vasopressin secretion in response to stimuli producing learned taste aversions in rats. Behavioral neuroscience. 1986;100:466-75.